Peer-to-Peer Support for Older Adults—What Do We Know and Where Do We Go? | Geriatrics | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Views 1,434
Citations 0
Invited Commentary
June 15, 2021

Peer-to-Peer Support for Older Adults—What Do We Know and Where Do We Go?

Author Affiliations
  • 1Lady Davis Institute for Medical Research, Jewish General Hospital, Montréal, Québec, Canada
  • 2Department of Psychiatry, McGill University, Montréal, Québec, Canada
JAMA Netw Open. 2021;4(6):e2113941. doi:10.1001/jamanetworkopen.2021.13941

As our population ages, there is an increasing need to develop effective, scalable interventions to support health and well-being among older adults. Scalable health care interventions are those that can be feasibly delivered to large numbers of people affected by adversity, including, for instance, communities with limited economic resources, people who are unable to access needed supports because of inadequate insurance or other barriers, or people confronted with natural or human-made disasters. Strategies to improve intervention scalability might include nonspecialist-delivered or self-help versions of evidence-based interventions, group-based interventions, and peer-support or peer-navigator interventions.1 Efforts to develop scalable interventions have been most prominent in mental health,1 but, among older adults, interventions have also been developed in other areas, such as to increase physical activity.2

There are few well-conducted randomized clinical trials (RCTs) on social support and other scalable interventions among older adults, which limits our ability to draw conclusions about intervention effects or relative advantages and disadvantages of different types of interventions.2,3 An overview of systematic reviews,3 however, found that processes and mechanisms that may facilitate successful implementation include bringing together people with similar experiences or characteristics and enabling participants to interact freely and develop meaningful relationships.

COVID-19 has pushed governments and other institutions that provide health care to seek scalable interventions to address multiple population-based needs, including mental health, and a living systematic review identified trials of 2 such interventions.4 One, a lay-delivered supportive telephone calling intervention, emphasized training of volunteers in empathetic listening and was effective in reducing symptoms of anxiety and depression among homebound older adults enrolled in a Meals on Wheels program in the US. The other, a peer-moderated group-based videoconference intervention, had similar effects among people from 12 countries with the rare autoimmune disease systemic sclerosis. These examples suggest that peer-to-peer interventions may be an effective solution when professional resources are overstretched, as has been the case during the pandemic, but more research is needed.

Schwei and colleagues5 report secondary results from a study they conducted between 2015 and 2017 that assessed peer-to-peer support provided to medically or socioeconomically vulnerable community-based adults aged 65 years and older receiving services from 3 social welfare organizations. Peer-to-peer support involved matching a trained older adult volunteer with a less able peer to provide a range of assistance, including home visits, transportation for shopping or medical appointments, and regular social connection. This was an observational study, and the amount and specific nature of peer-to-peer interactions were not prescribed but were tailored on an individual basis depending on the perceived needs of participants. A total of 456 participants were enrolled in the study, including 222 adults who were receiving peer-to-peer support and 234 adults who received standard community support only. Adults receiving services at study centers were eligible for peer-to-peer support if they were living in the community, were living at or below poverty level or on a fixed income and unable to meet living expenses, were socially isolated, had at least 1 chronic illness, or were frequent users of community organization services. Schwei et al5 enrolled participants receiving peer-to-peer support and participants receiving only standard services; services were not assigned as part of the study. Older adults in 2 centers were ongoing users of peer-to-peer support programming at the time of study initiation, whereas older adults in 1 center were new users at the time of study enrollment. For the study, participants receiving and not receiving peer-to-peer support were matched at each study site according to age, sex, and race/ethnicity, and analyses were done using propensity scores to attempt to control for confounding.

In their report of primary results from the study,6 the authors found that there was a statistically significantly higher rate of hospitalizations in the peer-to-peer support group (0.68 vs 0.44 hospitalizations per year; risk ratio, 1.54; 95% CI, 1.14-2.07) and no differences in emergency department or usual care visits. In their report of secondary outcomes,5 they found that participants receiving peer-to-peer support had statistically significantly worse anxiety symptoms and that there were no differences between groups in depressive symptoms or mental and physical components of quality of life.

From these results, one might conclude that peer-to-peer programs, including that studied by Schwei and colleagues5 and Jacobs et al,6 should be reconsidered; participants in the program did not appear to have benefited and, in fact, may have had worse outcomes. Such a conclusion, however, could risk abandoning potentially helpful, inexpensive programming without high-quality trial evidence on likely benefit or harms. We typically think about bias in our studies as leading to spuriously positive results and portraying health care interventions in a better light than is merited. Bias, however, is not unidirectional, and it can also lead us to draw unwarranted conclusions about lack of effectiveness.

There are 2 major reasons why we should not be too quick to dismiss peer-to-peer support programs on the basis of the findings reported by Schwei and colleagues.5 First, this was not a randomized trial. Study participants were identified and invited to enroll in peer-to-peer support programs on the basis of characteristics, such as income and social isolation, that would be expected to be associated with the outcomes studied. Because participants were initially enrolled in peer-to-peer support programs outside the study, other perceived risk factors could have also influenced to whom health care practitioners offered peer-to-peer support. Statistical controls were implemented in the study, but such controls cannot always account for imbalances, particularly in difficult or impossible to measure attributes that may be perceived and acted on by health care practitioners. It is possible that naturally, without intervention, people in the peer-to-peer group might have fared worse than those in the standard care group, although it is impossible to know whether this was the case and, if so, how much worse. Second, of participants in the peer-to-peer condition, only 16% were enrolled at the single site that was newly initiating a peer-to-peer program at the time of study enrollment. Most participants were already receiving peer-to-peer support before initiation of the observational study; if there was an initial benefit from peer-to-peer support that was then sustained, this would have been measured as no gain over the course of the study period.

Given the overall evidence base2,3 and recent successful use of lay-led or peer-led interventions during COVID-19,4 there is hope, albeit a great deal of uncertainty, that peer-to-peer interventions may be scalable, effective options to improve quality of life for older adults. Well-conducted RCTs are needed, however. Ideally, primary outcomes in such trials will be patient-important outcomes, related to quality of life. Outcomes that reflect health care service usage, such as hospitalizations or medical visits, are less important to patients. In addition, they are difficult to influence, even for highly targeted, professionally delivered programs. For instance, professional interventions to prevent falls in older adults reduce falls successfully but not hospitalizations, emergency department visits, or outpatient visits.7 Given that peer-to-peer interventions are designed for scalability and to meet unmet needs feasibly, we should not expect this from them either. Consistent evidence from well-conducted RCTs of even small benefits in quality of life would be enough, but trials are needed to establish whether peer-to-peer programs can deliver this.

Back to top
Article Information

Published: June 15, 2021. doi:10.1001/jamanetworkopen.2021.13941

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Thombs BD et al. JAMA Network Open.

Corresponding Author: Brett D. Thombs, PhD, Lady Davis Institute for Medical Research, Jewish General Hospital, 4333 Cote Ste Catherine Rd, Montreal, QC H3T 1E4, Canada (

Conflict of Interest Disclosures: None reported.

Funding/Support: Dr Thombs was supported by a Tier 1 Canada Research Chair. Ms Carboni-Jiménez was supported by a McGill University Faculty of Medicine Solvay Fellowship and by a McGill University Delta Upsilon Scholarship.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

World Health Organization. Mental health: scalable psychological interventions. Published August 2017. Accessed May 10, 2021.
Burton  E, Farrier  K, Hill  KD, Codde  J, Airey  P, Hill  AM.  Effectiveness of peers in delivering programs or motivating older people to increase their participation in physical activity: systematic review and meta-analysis.   J Sports Sci. 2018;36(6):666-678. doi:10.1080/02640414.2017.1329549PubMedGoogle ScholarCrossref
Boulton  E, Kneale  D, Stansfield  C,  et al.  Rapid systematic review of systematic reviews: what befriending, social support and low intensity psychosocial interventions, delivered remotely, are effective in reducing social isolation and loneliness among older adults? how do they work?   F1000Research. 2020;9:1368. doi:10.12688/f1000research.27076.1Google ScholarCrossref
Bonardi  O, Wang  Y, Li  K,  et al.  Effects of COVID-19 mental health interventions among community-based children, adolescents, and adults: a living systematic review of randomised controlled trials.   medRxiv. Published online May 7, 2021. doi:10.1101/2021.05.04.21256517Google Scholar
Schwei  RJ, Hetzel  S, Kim  KM,  et al.  Peer-to-peer support and changes in health and well-being in older adults over time.   JAMA Netw Open. 2021;4(6):e2112441. doi:10.1001/jamanetworkopen.2021.12441Google Scholar
Jacobs  EA, Schwei  R, Hetzel  S,  et al.  Evaluation of peer-to-peer support and health care utilization among community-dwelling older adults.   JAMA Netw Open. 2020;3(12):e2030090. doi:10.1001/jamanetworkopen.2020.30090PubMedGoogle Scholar
Tricco  AC, Thomas  SM, Veroniki  AA,  et al.  Comparison of interventions for preventing falls in older adults: a systematic review and meta-analysis.   JAMA. 2017;318(17):1687-1699. doi:10.1001/jama.2017.15006PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words